Indian Institute of Technology Kanpur

National Programme on Technology Enhanced Learning (NPTEL)

Course Title
Manufacturing Process Technology – Part- 1

Module- 25

by
Prof. Shantanu Bhattacharya

Hello and welcome to this manufacturing process technology part1 module 25

(Refer Slide Time: 00:17)

We were talking about the temperature distribution of a sand molding and one side of the
molding on the temperature boundary condition was that equal to the freezing temperature of the
material the other side it was open to the atmosphere and we were trying to decide for what
would be the distribution of θ per sale with respect to time and temperature. So for that we
actually done this similarity variable analyses yesterday.

(Refer Slide Time: 00: 41)

1
Through which we got a variable η and then try to convert the transfer problem in time and
space to the variable η okay and so what we found out was his expression dθ/dη=y=c2η 2 or just
writing that back here dθ/dη=C2e-η2 meaning there by that θ can be represented as if you just
integrate θ you know dθ it can be represented as integral 0 to η, e -z2 dz+c1 and let’s now decide
for what are C2 and C1 through the normal boundary conditions so if supposing η which is
actually x/2√αt for x tending to 0 also becomes 0.

So the η tends to 0 here and that there would be actually equal to C1 and at the boundary of the
mold where x is 0 you already know that the θ has been defined as freezing temperature θf. It
may just recall this mold was the something like sand mold okay on one side of it the boundary
was corresponding to the start of x=0 as x would proceed all the way to ∞ add x to ∞ and there is
ambient air which should be all round this particular mold okay.

So θ=θf here and the other thing would correspond to eventually when for the large amount of
time the θ would tend to be equal to θ 0 at x=∞ meaning there by that you know it is always sort
of it is actually equal to room temperature obviously it’s very, very large distance away so it
should be equal to ambient temperature so this is θ0 okay.

So having said that now let us now try to evaluate θf becomes equal C1 corresponding to η=0 at
this particular phase you know θ= θf here and therefore θ can be represented as C2 integral 0 to
η, e-e2dz+ θf

2
And let’s look at the case when η is ∞ that means corresponding to x trending to infinity when θ
becomes equal θ0 so you have θ0 here this is C2 integral 0 to ∞ e-e2dz+ θf and this is actually
integral whose value is estimated as √pi/2.

So that is how you know you can estimate what is going to be the C2 value for in this particular
case the C2 times of √Π/2 happens to be equal θ 0-θf and so the C2 would come out to be equal
twice by √Π ( θ0-θf ). so having found out what is C2 and what is c1 through these two equations
I substitute this back into the final form of the equation here which we call 7 and try to find out
what is the distribution of θ.

So the θ then becomes equal C2. let’s write the equation once more C2 integral 0 to η e- e2dz+ θf
which is actually the temperature boundary condition along the phase of the mold on c2 here
becomes equal to again if you just look back C2becomes twice by √Π ( θ 0-θf ). so twice by √Π
( θ0-θf ) times of integral 0 to η e-e2dz+ θf so that is how θ is represented on another words if I
just wanted to add and delete θ0 so basically let’s say we had and believe θ0 here.

So I would have much simpler form of the expression as θ 0+ θf - θ0 times of (1-2√ η) 0 to η e-

e2
dz and this is nothing but the error function of the term η okay and η basically is nothing but
numerical integral so η basically nothing but x/2√αt and so therefore the value of θ comes out be
equal to θ0+ θf - θ0 times of 1-error function of x/2√αt so that is how I would I to finally express
the θ calling it equation 8 and then you know this is the only case where I have in the interest of
just

(Refer Slide Time: 06:08)

3
Detail knowledge I try to prove what is the value of θ you have to understand the all other
condition that are trying to happen including the condition where there is resistance at the you
know the mold wall because of the thin air fill developing between the liquid metal and the mold
or in the case when there is a liquid to liquid or a solid to solid complete welding you know
because of a solid mold the only things which vary there are the boundary conditions.

So we will try to sort of evaluate and various boundary condition and decide what is going to be
the transfer equation in those cases analyse there is curvature etc these are the normal modality
which are followed, so that is how you basically represents the temperature distribution function
of the mold as the function of distance and here it is probably for me to define the error function
so the error function of the function z can be written down series as twice by √Π times of z-z
cube /3! so this is from common knowledge of mathematics I’m not going to go into the details
of all this but just to the present because we need these illustration of these sort of you know
proof the various boundary condition etc.

And then also there are function of z as you know can be written down as the 2 by root pi times
of integral of 0 to z e-x2 dx and obviously the value of this integral of this numerical integral is
commonly available in tables through which you would like to sort of illustrate or actually try to
find out what are the real times of solidification etc based on distribution of etc. so also that d/dz
of this error function should be able to get represented as 2√Πe -z2 so these are some things we
call it equation 9,10 and 11 it’s the some of the basic mathematical forms of the error function

4
that we would like to investigate or like to use basically later on from our analyse point of view
and we already know how this error function is related with respect to θ with the function of x.

(Refer Slide Time: 08:41)

[
θ x ( t )=θ0 + ( θ f −θ 0 ) 1− erf
x
2 √ αt ]
Q́=− kA ( ddxθ )
x

x=0
So this is an actual summary of whole form of representation of θ so θx is now the temperature at
a distance x from the mold phase at the distance t and α is the thermal diffusivity which is
normally given my thermal conductivity k/row c that is how it was defined in the heat transfer
equation the thermal conductivity, rho is the density of the liquid metal.

And C is specific heat capacity density of the mold I’m sorry and heat is the specific heat
capacity of the mold as well this temperature distribution mind you in mold so it’s actually the
sand properties that we are evaluating here while doing the θ distribution within the sand mold
and therefore now you already know that the rate of heat flow which would happen from the
metal to the mold would really depend on –k , the thermal conductivity times the area of the
mold interface with respect to the liquid metal times of difference ingredient with respect to x so
this dθ/dx is something that we have to look at boundary x=0 so you already know that the sand
mold as I illustrated earlier had a boundary where the θ value that was there was corresponding
to the freezing temperature.

5
And the heat transfer that was happening was typically in this direction across the area of the
mold metal interface so the amount of heat quantity now heat rate Q dot that is minted from the
obviously the metal which is at the higher temperature to the mold is actually is depend on –ka
dθx/dx at x=0. so that is how you basically try to decide for the Q dot or the rate of heat transfer
so having said that now the obvious idea that would come into picture is how to sort of calculate
what is this Q dot okay.

And try to then illustrate everything in terms of temperatures θ f, θ0 so on so forth so we have to

somehow now be able to find out mathematically what is this dθx/dx from this expression θx so
that it should be able to calculate what is the rate of heat flow, so let’s do that. so here you have
side dq/dt which is actually equal to –kAdθx/dx from this simple four year law add x=0 and the
θx as a function of x has been given out as θ 0 times of θf-θ0 times of 1-the error function of
x/2√αt and therefore the δθx/δx here comes out equal to θ f-θ0 with the minus sign of course times
of 2√Π times of you know 1/√πt e-x2/αt2.

So that is how you calculate the error function because already I have defined d/dz of the error
e-z2
function of z as equal to 2/√Π okay. So in this case z off course is x/2√αt and so the d/dz of
e-x/2√αt2 so
the error function of x/2√αt would actually become 2/√Π it is basically 1/2√αt so having
said having said that.
z3 z5 z7
erf ( z )= (
2
√π
z
z− + −
3.1! 5.2! 7.3 !
+… )
2 2
−x
erf ( z )=
√π 0
∫ e dx
d 2 −z 2

erf ( z )= e
dz √π
So that is how we substituted this value right about here so let’s now look at what really that
would be so this is actually the two goes on√ here and we are left with at x=0 if you want to
calculate the δθ x/δx x=0 that becomes equal to θ 0-θf so the minus is taken inside times of 1/√
ΠαtE0 which is actually can be again further written as θ 0-θf/√Παt1/2 so that is how dθx/d and
therefore the dq/dt that I was talking about –k times of a times of θ0-θf divided by again the in
another words it can be written down as k θf-θ0/Παt 1/2 so that is how the dq/dtx=0 is going to be
defined.

6
So having said that now the total quantity of heat flow across the mold phase up to a certain time

t0

t0 is qt0 the total amount of heat is Qt =∫ Q́dt

0
okay so if I really want to put this integral here
0

kA ( θf −θ 0 )
to solve the Q0 had earlier the Q dot and come out to be Q́= , so this value would
√ παt
t0
2 kA ( θ f −θ 0 )
then actually be equal to Qt =∫ Q́dt= √t0 .
0
0 √ πα
So again so t-1/2∫-1/t1 that is how you calculate this integral okay so that is how your total heat
from time 0 to t0 across the mold phase would actually flow okay. next calculate the heat that the
liquid metal would reject in order to solidify so if the liquid metal as latent heat l , a specific heat
capacity to be Cm and density ϼm then the heat rejected by the liquid metal converting to solid
can be written down as ϼm times of v obviously the mass times of the mass that has been
converted into solid times of the latent heat l which goes on detected because there is a phase
change times of Cm the specific heat capacity of the metal from the poring temperature to the
freezing temperature. So you assume that the solid that has been formulated is at the θ freezing
point or θ freezing temperature.

(Refer Slide Time: 17:51)

7
And the poring temperature is different than the freezing temperature meaning there by you are
super heating the material in order to induce into the mold just that the time of solidification can
be delayed a little bit and you don’t have any problems in filling the mold with the reverse flow
of the riser side, so that in that situation you having the total amount of heat reject Qr=ϼmv[L+
Cm (θP-θf)].
QR =ρm V [ L+ Cm ( θ p −θ f ) ]
So have said that now if I incorporate this into our expression earlier so the total amount of heat
rejected should be the total heat that flows out so we have an equation KA/θf-θ0 times of √Πα
times of √t0 that is the amount of time this heat would take for translating okay so this I actually
call t0 at this point and I will let u know what exactly this value means should be equal to
ϼmv[L+ Cm (θP-θf)].
2 kA ( θf − θ0 )
√t s= ρm V [ L+C m ( θ p − θ f ) ]
√ πα
So this time t0 is nothing but the total amount of time that the liquid metal would need to solidify
and so therefore we can record this as the solidification time okay so this is how you principally
in principal calculate this solidification time of a certain mass of a liquid getting converted into
solid because of the heat transfer across the mold wall assuming that there is a temperature
distribution with the function of x starting at the mold wall onwards into the particular sand
mold.

So having said that I think I have sufficient justification for you to be able to estimate now we
took at practical problem so for that we need to probably break here and then go back into the
next module later on thank you so much.

Acknowledgement

Ministry of Human Resources & Development

Prof. Satyaki Roy

Co – ordinator, NPTEL IIT Kanpur

NPTEL Team
Sanjay Pal
Ashish Singh
Badal Pradhan
Tapobrata Das
Ram Chandra
8
 Dilip Tripathi
Manoj Shrivastava
Padam Shukla
Sanjay Mishra
Shubham Rawat
Shikha Gupta
K.K Mishra
Aradhana Singh
Sweta
Ashutosh Gairola
Dilip Katiyar
Sharwan
Hari Ram
Bhadra Rao
Puneet Kumar Bajpai
Lalty Dutta
Ajay Kanaujia
Shivendra Kumar Tiwari

an IIT Kanpur Production

@copyright reserved